Each somatic cell in a human body contains the same genetic information, but the diverse function of differentiated cells is made possible through precise control of gene activity. Proper gene regulation is crucial for the normal function of a cell, and human diseases can and do result from loss or circumvention of normal gene regulation. Our long term objective is therefore to understand how gene expression is regulated in eukaryotic cells. This grant proposal focuses on transcriptional coactivators which enable the TATA-binding protein (TBP) to bind to its TATA-box DMA binding site, a critical step in initiating gene transcription. Significant data indicates that the SAGA histone acetyltransferase (HAT) complex functions as a coactivator by delivering TBP to the promoter and by acetylating nucleosomes in the promoter region. Despite major advances in the last decade, the molecular mechanism for either activity is not well-understood. For example, we do not know precisely how SAGA recruits TBP to the core promoter. Similarly, while we do understand how HAT subunits recognize histone tail substrates, we do not have an equivalent understanding of how HAT or other chromatin enzymes recognize their physiological nucleosome substrate. Our specific aims are directed to address these deficiencies. In particular, we propose to: 1. Determine how SAGA's Spt8 subunit binds to TBP. We will perform mutagenesis and label transfer experiments to define the TBP and Spt8 surfaces that interact. 2. Determine how SAGA recruits TBP to a promoter TATA-box. We will develop and utilize in vitro assays to examine the molecular mechanism for how SAGA delivers TBP to a promoter. 3. Determine how the Ada2/Ada3/Gcn5 and Piccolo/NuA4 HAT enzymes interact with nucleosomes. We will use photoaffinity labeling to identify HAT complex subunits in close proximity to nucleosomes, and deletion and mutagenesis studies to define novel functions for the bromo and chromodomains in the binding and acetylation of nucleosomes. Our biochemical studies using defined components will complement the considerable available in vivo data by addressing mechanistic questions not easily tackled by other approaches.